· web viewfor example, the common title “title vii” refers to title vii of the civil rights...

ELIS/GLIN XML DTD Annotated April 18, 2001 p. 1

Copyright 2001 Konstantinos Kalpakis, Steven D. Jamar, and Kenneth J. Markowitz

Annotated XML Legal Document DTD for ELIS & GLIN

by

Prof. Steven D. Jamar, Dr. Konstantinos Kalpakis, and Kenneth J. Markowitz, Esq.

April 18, 2001

Executive Summary

The Environmental Legal Information System group has created a XML document type definition (DTD) to be used by ELIS and to be submitted for use by GLIN and ELINE, should they so desire, for legal documents to be entered into their respective databases.

In crafting the DTD we balanced the following needs and aims:

1. The DTD should be sufficiently simple so that it can easily be understood;

2. The DTD should have sufficient power to allow appropriate manipulation, sorting, display, retrieval, and indexing of documents;

3. The DTD should be general or generic enough that nearly any legal document, ranging from local ordinances to international treaties, and including court or tribunal decisions as well as journal articles and other legal publications, can be tagged and entered into the databases;

4. The DTD should be useable for laws from many countries in many languages;

5. The DTD should be flexible enough to allow modifications to it for special needs without requiring complete reworking of it.

6. Attention has been paid to issues of authenticity and archiving, but this DTD is not structured to insure those important objectives by itself. Such concerns will need to be addressed by those undertaking to develop and maintain authentic, accurate legal data. ELIS proper is not undertaking that function.

Although this DTD is intended to be used for a wide variety of legal documents, it is not intended to be used without additions to it for contracts, court filings, and legislative bills in process (tracking amendments, and the like). Nor does it include publishing markup of the sort which may be required by the GPO. And it does not include a means to automatically update and revise statutes as they are amended. In a sense, this DTD is intended for marking a law as a snapshot. Historical functionality can be added to it through extended DTDs, but this DTD is not aimed primarily at that function.

In short, our intention is to capture the information about the document in a way that permits it to be entered into and retreived from a database and to be displayed in a useful, readable, reliable, durable format. Concerns about searching the documents and indexing are also addressed.

This DTD permits the capture of necessary information to uniquely identify the document and the capture of certain critical information about it such as the source (e.g., country,



municipality, or international organization), author, date of entry into force (for treaties), and the like. Although it does not attempt to capture all information that could be relevant, it does permit indexing through use of a key word thesaurus.

The basic unit contemplated by the DTD varies with the type of document. For court or tribunal decisions or opinions the basic unit is the entire text of the opinion. There may be options added for case opinions which permit tagging facts, rules of law, quotations of other sources of law within the opinion, etc., but these are not part of the main DTD. For statutes, ordinances, and longer treaties, the basic document unit is the separately indentified paragraph or section (generically understood – these could be called clauses, articles, or some such). The data for statutes and statute-like documents is captured in such a way as to allow the statute to be rebuilt or to allow display or retrieval of just a section.

The main elements and their general meanings are as follows:

1. LegalDocument: This is the parent element, or the main “container” for all of the other elements. Legal documents can be tied to one another through the table of contents and through cross references. Thus a legal document can be a part of a larger document, and, with respect to statutes and treaties, typically should be.

2. DocID: A unique document identification number for the document for the particular database. This will typically be assigned by the data entry program or by the program loading the document into the database. This is not intended as a universal identifier.

3. Country: The originating country is captured using the international 2 character code. A dictionary will be used to translate back and forth between full names, other abbreviations, and the two digit code as well as to track changing names of countries. This element is used more than once and the exact meaning depends on the context.

4. LegalDocumentClass: This element is used to capture information about the nature of the source of the document (national, local, international, regional governmental body or non-governmental or inter-governmental organization) and the kind of document it is, e.g., treaty, statute, executive order, tribunal or court decision, commentary, etc. This information is useful because type of information to be captured and the way it is to be displayed will vary by instrument classification.

5. Title: This element is used for display, searching, and citation. Official and common and shorthand titles can all be captured as can titles in multiple languages.

6. Summary: The summary is an abstract or other summary of the document. For GLIN, this would include, among other things, English language summaries of laws published in other languages.

7. TableOfContents: The Table of Contents is generated automatically and is used to link related legal documents to each other.



8. LegalDocumentText: This element is to tag the text of the document itself.

9. EffectivePeriod: This element is used to indidate the start date, and where appropriate, the end date.

10. Subject: One key way to find relevant information is by the subjects the document addresses. This element allows for the capture of subject metadata which in turn will allow topical searches and will allow for tying documents to other types of information on the same topic, e.g., remote sensing data and GIS/digital earth data.

11. GeographicDomain: We are advocating having laws geo-referenced. This element captures geo-referencing information such as where the law is effective, e.g., a particular park, watershed, country, or other political subdivision.

12. Citation: This captures the information to uniquely identify the document, to identify its location (typically in hard copy), to permit linkages to other online resources, and to allow creation of standard citations automatically.

13. CrossReferences: This captures information which links the document to another document. It may also be used for citations within the document to other documents.

14. Methodology: This element would be used for secondary sources such as studies of the efficacy of a particular treaty or program. The researcher would disclose the methodology used and it would be separately tagged.

15. CaseInfo: This element permits inclusion of information typical of court or other tribunal decisions including information identifying the tribunal, the date of the decision, and the names of the parties and attorneys.

16. TreatyInfo: Most of the information about a treaty is captured by the regular document information, but this allows specific tagging of additional information such as date of opening for signature, entry into force, and ratification by a particular state.

17. Organization: This element captures information about organizations related to the legal document in some way. An organization can be the author, enforcement body, regulated group, watchdog organization, party to a treaty, and so on.

18. AccessConstraints: This permits restrition of access to those with proper passwords or opens it up to a wider group.

19. RecordInfo: This is used to tie various related sections of a treaty or other document together so that they can be recreated properly and displayed in various ways including single sections or multiple sections or parts.

This DTD does not include any style information. That will be handled by style sheets using the tags to present the information.

ELIS/GLIN General Legal Document DTD

The basic unit contemplated by this DTD for most legal documents other than case decisions is a separately identified paragraph or section. For case decisions or opinions the basic unit is the entire text of the opinion. There may be options added for case opinions which permit tagging facts, rules of law, quotations of other sources of law within the opinion, etc., but these are not part of the main DTD.

The comments are intend to permit the reader to understand and implement the DTD more easily. Other materials will be created describing marking up documents and using appropriate xml data entry forms as those are developed.

The DTD has several parts. (These parts have no meaning within the DTD itself, but they may make understanding it a bit easier.) The first part defines entities to be used for data entry for some of the elements. The second part includes basic defining information about all legal documents including such items as title, summary (or abstract), and table of contents (and links to other parts of the document). Then comes the third part, the element for the text of the document itself. After the text element additional information about that particular text is captured in the fourth part of the DTD - information such as date information, subject matter, geo-referencing, and citation. In addition, for cases and treaties additional specialized information is captured. The fifth part of the DTD captures miscellaneous other information about things related to the legal document such as organizations involved, user access constraints, and various record information including links to pdf or other files of the same document.

<?xml version="1.0" encoding="US-ASCII" ?>



The entities below define permissible values for the attributes of some elements. The LegalXML group (an organization seeking to set open standards for xml implementations in legal settings) convention of naming is being followed, i.e., semantically meaningful names, few abbreviations, and capitalizing the initial letter of each word in the name. In order to fully understand the entities, the contexts in which they are employed need to be understood. The following definitions and explanations should be read in conjunction with the actual use of the entities in the DTD elements. A table will be created to cross-reference entities with elements and elements used in more than one part of the DTD.These entities serve as macros to limit the meaning of certain attributes. This set of entities is checked when processing certain elements to help insure uniformity and appropriate content for a number of the elements. Use of entities which are used more than once also streamlines the DTD.



<!ENTITY % TitleType "(official|common|citation|other) 'official'" >This entity is used in the element “Title,” an element of the main container, LegalDocument.This DTD allows documents to have many different titles. Statutes are known by official names, common names, acronyms, and the like. In addition, certain documents may have official versions with official titles in several languages, e.g., U.N. treaties. The “TitleType” identifies whether the title being used for the particular document is the official or common or popular or other sort of title. This information is relevant for how the document is presented and may be relevant for the citation. This entity “TitleType” limits the selection of type of title to four types with the default type being “official.” Limiting selection like this insures that the data will be entered correctly (no misspellings), or at least that it will be entered in a way which the software can handle. For example, the common title “Title VII” refers to Title VII of the Civil Rights Act of 1964. The Civil Rights Act of 1964 would be a popular name or common title, as would Title VII. Acronyms like ERISA and ERTA would also be “common” names or could be “other” names.

Official: The official name given to the legal document upon creation of it. The title will vary substantially depending upon the kind of document, e.g., treaty, tribunal decision, or statute.

Common: A title by which the legal document is commonly known. This would be of particular relevance for treaties or statutes.

Citation: The title used in citation if different from the “Official” citation.Other: This would be rarely used. There may be instances where a title is neither official

nor a common or popular name, but which is used to refer to the legal document. <!ENTITY % Language "(en|sp|fr|de) 'en'" >

This entity is used for the elements Title, Summary, table of contents element ContentsEntryText, PrefatoryStatement, LegalDocumentText, Subject (for indexing), GeoName (for geo-referencing), Citation, Treaty ActionDetails. This entity limits the method of specifying the language used for a document or segment of the document to standard international two-digit codes. All valid ISO codes for all the languages of the world that have codes are accepted. Only four are shown for the sake of brevity and illustration. The default language is English.

<!ENTITY % CountryCode "(us|eu|ca|mx) 'us'" >This entity is used for the element Country for basic address information for elements like Party, Counsel, and Organization.This entity limits the method of specifying the country or organization, e.g., European Union or the United Nations, to standard international two-digit codes. All valid ISO codes for all the countries of the world are accepted. Only four are shown for the sake of brevity and illustration. This entity is used whenever country data is required or allowed, e.g., for source of the document or for contact addresses, etc. The default country is the USA. For entry forms when entering multiple documents, the element which uses this entity can be set to enter the right country by default.

<!ENTITY % ContentLevel "(1|2|3|4|5|6) '1'" >



Specifies level of the document with respect to other related legal documents. For example sections of a statute would be on one level with the article or chapter being above it. This will typically be computed, but can be entered manually. This is used in part to help organize and link documents together. It is explained more fully where it is used in the TableOfContents element.

<!ENTITY % ThesaurusCode "CDATA #IMPLIED" >A thesaurus gives unambiguous meaning to terms which permits higher quality indexing and searching. This entity is used to unambiguously identify the proper thesaurus to be used. The default value is empty. Each document can have default thesaurus or a variety of thesauri can be specified. The thesaurus does not need to be one created by ELIS or GLIN or anyone else using this DTD; it can refer to any thesaurus. If a thesaurus is published and given a code, that code can be used here to unambiguously identify that thesaurus.

<!ENTITY % SubjectTermType "(controlled|uncontrolled) 'uncontrolled'" >This is used for indexing and searching. It is used whenever specifying a subject code or keyword. The term can come from a thesaurus or it can be a free text keyword. “Controlled” refers to string defined in a thesaurus. “Uncontrolled” means that the term is not from a thesaurus and simple string matching is used. Using Because avoids two subject words can have identical meaning. So can have two terms that make things equivalent. Controlled would change CO to carbon monoxide in one thesaurus or Colorado in another or just CO as a string if uncontrolled. If two thesauri exist, then you need to specify which thesaurus - e.g., environmental or states.

<!ENTITY % CitationType "(primary|alternative|universal) 'primary' " >Many sources have multiple citations including “universal citations” or vendor-neutral citations and the more common book-based citations. Some items are found in multiple books (e.g., the U.S. Code or the U.S. Supreme Court Opinions) as well as online. This element which uses this entity will be used for all citations and will differentiate the official citation (the primary citation) from other citations since there can be multiple citations.

Primary: This is the citation to the official citation source according to whatever system of citation is followed, e.g., AWLD Citation Manual.

Alternative: This is used for parallel cites to the same thing, including, for example, online sites for print materials.

Universal: This is used for vendor-neutral or “universal” citation if it is different from the primary citation.

<!ENTITY % PartyActionOnTreaty "(draft|approve|ratify|sign) 'draft'" >Classifies the action of a participant in a treaty. It is not about the treaty per se, but about a party’s relationship to the treaty.

<!ENTITY % OrganizationRole "(originator|author|contributor|distributor|implementor|other) 'other'" >

This entity is used in the element “Organization” to identify the role of the organization with respect to the particular document. A number of organizations (or persons) can play a variety of roles with respect to the particular document. With respect to a statute, for example, an organization could be the originator (e.g., the U.S. government as a whole), while another organization could be



charged with implementing the law, e.g., the EPA. Contributor can be used to identify from whom the particular legal document was received, e.g., the GPO, or some third party. For example, a person other than a national government may contribute that state’s constitution to be loaded into the database. The terms have the following meanings:

Originator: Organization from where the document originated, e.g., UN, US, Brazil, etc.

Author: The author of the document, e.g., a judge of a court opinion. For certain documents the originator and the author may be the same.

Contributor: A person or organization that contributed the document to the database if different from the originator.

Distributor: A person who distributes the document. This can be an archivist, database owner, publisher, or other sort of distributor. For example, Thompson Publishing is a distributor of court decisions by the U.S. District Courts.

Implementor: A person who is charged with enforcing or otherwise administering the document, e.g., a committee or body created by a treaty, a department or agency in a government, and so on.

Other: This can be anyone associated with the document in some significant way, including NGOs, IGOs, educators, etc.

<!ENTITY % OrganizationLevel "(local|state|province|regional|national|international|other) 'other' " >

This entity is used in the element “Organization” for the attribute describing the level of the organization. The terms have the following meanings:

Local: This refers to any level of government below the state(for the United States) or provincial level including municipalities, cities, counties, school boards, etc. For non-state organizations, it refers to the lowest level of rule-making or decision-making authority.

State: For the United States, this refer to the constituent states and analogous political subdivisions such as the District of Columbia, Guam, and Puerto Rico.

Province: For most nations this refers to the subdivision immediately below the national level, regardless of whatever that subdivision may be called.

Regional: This refers to bodies greater than provinces, but less than the full country. For example, a country may have something called “the Great Lakes Regional Authority” or some similar authority which has jurisdiction over a region which comprises a number of provinces or parts of provinces.

National: National refers to things at the level of the full country or nation. A synonym is “federal.”

International: This refers to all organizations which exist between and among states including global organizations like the UN and WTO as well as regional organizations like EU and NAFTA.

Other: If the organization does not fit into one of the other categories, or if the status is uncertain, then use this one.

<!ENTITY % OrganizationType "(government|private|public|ngo|igo|other) 'other' " >This entity is for the element “Organization” and allows the capture of information about the type of organization. Where more than one term applies,



either use only the most specific term, or enter all that apply. The terms have the following meanings:

Government: Any official governmental organization.Private: Any private organization such as a private business.Public: Any public organization membership in which is generally open to the

public. NGO: A non-governmental organization, which often is, but need not be also a

public organization.IGO: An intergovernmental organization.Other: An organization or person who does not fit within one of the other

categories.

<!ELEMENT LegalDocument (DocID?, Country?, LegalDocumentClass?, Title*,Summary*, TableOfContents?,EffectivePeriod?, Subject*, Citation*,CrossReference?, Methodology?,GeographicDomain?, CaseInfo?, TreatyInfo?,PrefatoryStatement?, LegalDocumentText?,Organization*, AccessConstraints?DocumentImage*,RecordInfo?) >

“”LegalDocument” is the root element under which all of the document and metadata about the document are entered. It can be any kind of legal document – statute, treaty, decision, regulation, contract, pleading, commentary, summary, etc.The child elements are defined below. In general, this DTD captures information about the source of the document, about the subject matter of the document, the relationship of the document to other documents, the text and other contents of the document (e.g., charts, tables, and other graphics), and access privileges to the document. In addition information concerning citation of the document and geo-referencing and more is also captured.(The question mark means the data is optional. An asterisk means both that the data is optional and that the element may be repeated.)

<!ELEMENT DocID (#PCDATA ) >Unique identifier for each legal document assigned by the person entering it or by the program used to enter the data.



This embedded comment identifies the information that goes into classifying legal documents.

<!ELEMENT LegalDocumentClass (#PCDATA) ><!ATTLIST LegalDocumentClass system CDATA #IMPLIED ><!ATTLIST LegalDocumentClass type CDATA #IMPLIED ><!ATTLIST LegalDocumentClass subtype CDATA #IMPLIED >

LegalDocumentClass is an element with three attributes. The LegalDocumentClass element defines what sort of document the document is, e.g., treaty, statute, case decision, etc. This matters for which elements should be completed, for display purposes, and for interpretation.

System: Refers to the jurisdictional authority, e.g., US, some other state, the WTO, the UN, the ICJ, the EU, etc..

Type: Refers to the nature or type of the document, e.g., regulation, statute, court decision, executive order, treaty, etc. .

Subtype: Refers to a subclassification, if any, such as public law or private law or appropriations; bilateral treaty or multilateral treaty, etc.

<!ELEMENT Title (#PCDATA) >

<!ATTLIST Title type %TitleType; > <!ATTLIST Title language %Language; >

“Title” is an element with two attributes. The attribute “type” uses the entity “TitleType”, which includes the official title, common name, or other name, such as a popular acronym. The “language” attribute also refers to an entity, “Language” which is used to articulate what language the title was written in. There can have multiple official titles, possibly in different languages (especially common for treaties).

<!ELEMENT Summary (#PCDATA) ><!ATTLIST Summary language %Language; >

We are using “Summary” to refer to various types of abstracts and summaries. (The Dublin Core defines Description as follows: “Description may include but is not limited to: an abstract, table of contents, reference to a graphical representation of content or a free-text account of the content.” http://uk.dublincore.org/documents/1999/07/02/dces/. Under the Dublin Core a table of contents is treated as a description of the content of the resource. We are using Table of Contents quite differently.)

<!ELEMENT TableOfContents (ContentsEntry*) >The table of contents is a list of the entries in the document. It is generated by the program handling the data; it is not a separate table entered as such. Nonetheless,



a table of contents from a book or other document could be included as a legal document, but that table of contents would be static, as in print, not dynamically generated as under this DTD. That table of contents should be entered as a separate summary using the Summary element. Understanding how this part of the DTD works is central to understanding how the documents are being handled. Under this DTD, the legal document can be of any size or configuration. It can be text, or image, pdf files, or any other object. In concept the base unit of a legal document changes depending upon the type or classification of the document, but how the DTD handles the document does not vary in a material way based on how a document is parceled.For example, let’s start with a complex statute. A statute can have a number of levels of hierarchical structure including sections, subsections, paragraphs, subparagraphs, sub-subparagraphs and so on, getting smaller. It also can have sections, articles, parts, chapters, titles, and so on, getting larger. Furthermore, there is no standard terminology for the levels of the hierarchy and there is no standard number of levels. Under this DTD, the number of levels and the labels attached to those levels do not matter. All that matters is the relationship of one document to another as either a parent, child, or sibling. Assume that the main unit of a statute, or the smallest unit that you want to separately encode, is deemed to be a section. That section would be entered as a separately, uniquely identified document. All of the information about that section would be encoded using this DTD, including the full text proper of the section.Next assume that you did this for all 5 sections comprising the article (the next larger unit). You would now have 5 legal documents coded under this DTD. To tie them together and to show that they are part of the same article, a new, 6th legal document would be created. That document would consist essentially only of a title and links to these 5 sections. After entry of 2 or more articles within a chapter (the next larger unit), another legal document would be created which would link to the appropriate articles. The articles would carry with them the sections.The table of contents would thus consist of the items at the levels immediately below that where one is looking, and can include levels above and collateral to the one where one is looking if the display and handling software is so programmed.The virtues of this system of encoding documents are that it is simple, iterative, relative, and very plastic. If one were to want to encode a statute with the basic unit being subparagraphs (a tax code, perhaps), the same process can be followed without change. If one were to want to encode the whole article (e.g., to get older data into the system faster), that too would work.Furthermore, this flexibility allows appropriate handling of a variety of disparate documents. Case decisions, treaties, statutes, constitutions, and executive orders may all be best handled differently from one another with respect to what is the most logical “unit” of the document.This method also can make data entry easier, with the right data-entry tools which “know” whether you are in the same level or moving up a level or so on and so



can automatically do the linkages and enter the unchanging data from document to document.

<!ELEMENT ContentsEntry (ContentsEntryText, PageRange?, Linkage?) >ContentsEntryText: The text that appears in table of contents.PageRange: Refers to the pages in hard copy. This is for citation purposes and cross-

reference purposes. Archivists and those concerned about data integrity and identity of text will want to be able to check the text against any official, hard-copy version.

Linkage: Provides a way to connect the table of contents entries to other documents or to anchors within the document. XML linkages can be very powerful.

See the Citation element for more concerning PageRange and Linkage. They have the same meaning there as here. Their exact meaning and usage is contextual.

<!ATTLIST ContentsEntry level %ContentLevel; >ContentsEntry level is a single number which says “go this deep”. So it must be an Arabic numeral – not a letter or roman numeral. This number comes from the entity ContentLevel. It tells the document how many levels of children it has. This number is entered automatically as you make the links to the child documents.

<!ATTLIST ContentsEntry order CDATA #IMPLIED >This attribute is to allow the correct ordering of linked sections. Although typically the order should be created properly automatically, this allows the programmer to overrule the order of appearance within the table of contents itself if necessary.

<!ELEMENT ContentsEntryText (#PCDATA) ><!ATTLIST ContentsEntryText language %Language; >

This element is to be used to specify the contents entry and to specify the language the contents entry. The contents entries will often be taken from the titles of child documents.

<!ELEMENT PrefatoryStatement (#PCDATA) ><!ATTLIST PrefatoryStatement language %Language; >

This element allows the separate capture of preambulatory text which is not part of the operative part of the legal document itself. This may or may not be used. For many documents it would be empty. For example, if the basic unit of a statute were treated as a section, and since most sections do not have preambles, most documents would not have preambles. However, at the chapter or article level there may well be a preamble. In that circumstance, the chapter-level legal document may have only a title, preamble, and collection of links (within the contents element).

<!ELEMENT LegalDocumentText (#PCDATA) ><!ATTLIST LegalDocumentText language %Language; >

This element is for entry of the text proper, or object proper. The only attribute included here is language. Nothing in this DTD prevents other people from creating sub-elements to tag information within this element. For example, embedded citations, quotations from other sources, and key word/subject coding could all be sub-elements, even though they are not provided for within this DTD.

<!ELEMENT EffectivePeriod (TimePeriod*) ><!ELEMENT TimePeriod (FirstDate?, LastDate?) ><!ELEMENT FirstDate (#PCDATA) ><!ELEMENT LastDate (#PCDATA) ><!ELEMENT Date EMPTY>

<!ATTLIST Date Day CDATA #IMPLIED ><!ATTLIST Date Month CDATA #IMPLIED ><!ATTLIST Date Season CDATA #IMPLIED ><!ATTLIST Date Year CDATA #IMPLIED >

For many legal documents, including most cases and statutes and treaties, only the first date would be entered as the date the legal force of the item came into effect. For some documents, like multilateral treaties, additional date information is permitted to be entered below.The element “Date” is defined here to allow entry of numerical date information such as 12 for the day, 8 for the month, 2001 for the year, or textual information such as Winter or March. This form of date information is important for citation purposes.

<!ELEMENT Subject (Term*) >This element is used for encoding documents with subject or topic information. Such encoding is necessary for efficient, effective research. Subject coding can be either from a thesaurus, or from several different thesauri, or free-form key words not tied to a thesaurus. The attribute “Subject type” is used to indicate whether the term is “controlled” or “uncontrolled” and uses the entity SubjectTermType for this purpose. (The term “Subject” is being used to conform to the Dublin Core elements designed for citation to resources generally. http://mirrored.ukoln.ac.uk/dc/documents/rec-dces-19990702.htm.)

<!ATTLIST Subject type %SubjectTermType; ><!ATTLIST Subject thesaurus %Thesaurus; ><!ATTLIST Subject language %Language; >

<!ELEMENT Term (#PCDATA) ><!ATTLIST TermID CDATA #IMPLIED >

A “term” is a entry in the subject index. A term could be superfund site, wetland, deforestation, etc. This element will be very important for tying law to environmental indicators and then tying them to the appropriate digital earth information or satellite images.The coding for subject would look like this:

For “uncontrolled” terms: <Subject> <Term> computer</Term><s/Subject>

For “controlled” terms (those from a thesaurus): <Subject Thesarus= “A1”> <Term id=3 /> </Subject>

<!ELEMENT GeographicDomain (place*) >

We want all of the laws to be geo-referenced, though this is not necessary for use of this DTD. It is contemplated that a place can be specified by either a polygon method or by a name which refers to a pre-defined polygon. As georeferencing becomes more commonplace and standardized, we anticipate that using

geographic names and looking up the polygon relating to that name will be effective.

<!ELEMENT Place (GeoName*, (GeoPoint|GeoPolygon)* ) ><!ELEMENT GeoName (#PCDATA) >

<!ATTLIST GeoName thesaurus %Thesaurus; ><!ATTLIST GeoName language %Language; >

<!ELEMENT GeoPolygon (GeoPoint*) ><!ELEMENT GeoPoint EMPTY >

<!ATTLIST GeoPoint lat CDATA #IMPLIED ><!ATTLIST GeoPoint long CDATA #IMPLIED ><!ATTLIST GeoPoint alt CDATA #IMPLIED ><!ATTLIST GeoPoint units CDATA #IMPLIED ><!ATTLIST GeoPoint system CDATA #IMPLIED >



Immediately below is the Citation element. This element captures information on how to cite the legal document itself. These same elements can also be in cross references, or embedded citations. The aim is to provide the information sufficient to uniquely identify the document and to identify generally where it can be found. The terminology tracks with the Dublin Core, hence the use of the term “Creator” instead of “Author.” This was done in order to make referencing legal documents easier in more generalized Dublin-Core compliant xml information indexing systems.

<!ELEMENT Citation (Creator*, Title?, Editor?,Publication?, Volume?, Number?, Date?, Publisher*, PageRange?, ParaqraphRangeNote?Linkage*URL) >

<!ATTLIST Citation type %CitationTypes; >CitationType is an entity which defines whether this is the primary or alternate or universal citation. There typically would be one primary citation with potentially multiple alternate citations. Normally the primary citation will be to the official print publication. There may not be a primary citation for a particular document. This situation can arise where there is no accepted primary print publication or where there is no accepted standard citation method. In some instances there are

multiple “primary” citations, e.g., one for print and one for electronic, or in multiple languages. With the advent of universal or vendor or location-neutral citations, there could well be a universal citation form which might in time become the primary form with all other forms becoming alternate forms. For now we are including a specific type labeled “universal” so that machines can know how to handle it since in theory, at least, universal types should be more regular than other types.

<!ELEMENT Creator (#PCDATA) >

The Dublin Core calls the author the “creator.” Though we preferred “author,”, we have nonetheless adopted the term “Creator” to conform to the more general citation format gaining acceptance across a broader range of material.

The element “Title” has been previously defined and needs to be entered only once. In an entry form it is likely that all citation information, including title, would be in one place on the form. “Title” simply refers to the name by which the thing is known. A problem is how to handle journals or anthologies. The term “title”, as used here, refers to the writing created by the author of the particular piece. For the anthologies and journal title, e.g., Journal of International Law, the element “Publication” should be used.Note that the title to a case should normally be entered here even though parties will be entered below under the special information elements for cases.

<!ELEMENT Editor (#PCDATA) >

Anthologies published in book form typically have an editor (or editors) for the book and title for the book separate from the authors of the articles and titles of the articles collected in the anthology. This element allows one to distinguish between the two unambiguously. The term “Publication” below does double duty for both periodicals and anthologies.

<!ELEMENT Publication (#PCDATA) ><!ELEMENT Volume (#PCDATA) ><!ELEMENT Number (#PCDATA) >

Note that the element “date” is a subelement of citation. “Date” was defined above under element “LeglaDocumentText.” Date for quarterly journals could be entered as publishers indicate, e.g., “Winter,” or “First Quarter” and so on.

<!ELEMENT Publisher (#PCDATA) ><!ELEMENT PageRange (first?, last?) >

Pages are handled for citation purposes as a first and last page. This citation element is for the document itself, i.e., it discloses where in the hard copy this

particular legal document is found, e.g., pages 33-56. The element CrossReference is used for citations within the document to other documents.

<!ELEMENT ParaqraphRange (first?, last?) >Paragraphs are handled for citation purposes as a first and last paragraph. This citation element is for the document itself. Universal citation formats, as well as decisions of international tribunals, use numbered paragraphs as the normal way of publishing and locating information. Numbered paragraphs free the document from page numbers which vary from printing to printing and which are not medium neutral. The element CrossReference is used for citations within the document to other documents.

<!ELEMENT Note (#PCDATA) >This element allows the citation to be annotated with any necessary explanatory information.

This element permits linking this document to other documents, but as used here, it is intended to link to official sources of the document or to other published, electronic versions of the same document (or of a document which contains it).The Linkage element is used in a number of places in this DTD.

<!ELEMENT Methodology (#PCDATA) >

This information is for secondary sources such as studies of effectiveness of particular laws or of compliance with regulations or treaties. It is used to describe the methods used for analysis

<!ELEMENT CaseInfo (DecisionDate?, Party*, Counsel*, Judge*) >

The CaseInfo element allows specialized information to be captured for written decisions of courts or other tribunals. The case will have a date on which it was issued, one or more parties, and will probably have one or more attorneys.

<!ELEMENT DecisionDate (#PCDATA) >

The DecisionDate element allows for uniquely tagging the date the case was decided. The date format will be in standard ISO 8601date format yyyy-mm-dd. The decision was to use a special name for the decision date to distinguish it from any other dates which may be captured. This is one of the few decontextualized elements we have used.

<!ELEMENT Party (Name*, Organization?) >

<!ATTLIST Party ID ID #IMPLIED >“Name” is the name of an organization or a person.The Party element allows for uniquely identifying each party to the case. This element does not include breakdown by parts of names, though such information could be added and could be handled by if such tagging were on incoming data.

A unique identifier for each party is added by the software. Once this element is set by the Legal XML initiative, we may need to adjust our structure here.By using subelements “Name” and “Organization,” all of the information and structure from those elements are included within “Party.” Those elements are defined below under the Organization element.

<!ATTLIST Party role “(plaintiff|defendant|other) ‘other’”>The Role element is used to identify as plaintiff, defendant, or other role (e.g., intervenor).

<!ELEMENT Counsel (Name*, Organization?, Role?) >

By using subelements “Name” and “Organization,” all of the information and structure from those elements are included within “Party.” Those elements are defined below under the Organization element.

<!ATTLIST Counsel Client IDREFS #IMPLIED >This just ties the client/party to counsel.

<!ELEMENT CrossReferenceList? (CrossReference*) >The CrossReferenceList element is used to allow documents to refer to each other. At present it is only for basic linking of documents, but it could be redefined to include the ability to tag citations within the document to other documents. For such a citation, care should be taken to include the same information as in the Citation element, to comply with ALWD, to comply with medium-neutral citation forms, and to permit short forms and signals.

<!ELEMENT CrossReference (Title?, Citation*, Linkage?)* ><!ATTLIST CrossReference Relationship* CDATA #IMPLIED>

This attribute defines the relationship of this legal document to the one to which it is linked. Typically this would be a summary, commentary, amendment, etc.

<!ELEMENT Linkage (#PCDATA)><!ATTLIST Linkage Type CDATA #IMPLIED >

Type of object being linked to, e.g., pdf file, or html file, etc.<!ATTLIST Linkage Code CDATA #IMPLIED >

How to get to where you want to get and how to present it to the use. The “Code” could be ftp, http, anchor, gopher, rdf.

<!ELEMENT TreatyInfo (DraftedDate?, EffectiveTimePeriod?, TreatyAction*) >

This TreatyInfo element allows the entry of information specifically related to treaties such as date drafted, date entered into force, which states are parties to it, what action with respect to a treaty this document represents (e.g., ratification).

<!ELEMENT DraftedDate (#PCDATA) ><!ELEMENT EffectiveTimePeriod (FirstDate?, LastDate?) ><!ELEMENT TreatyAction (Date?, Party*, ActionDetails?) >

<!ATTLIST TreatyAction Type %TreatyActionTypes; ><!ELEMENT ActionDetails (#PCDATA) >

<!ATTLIST ActionDetails Language %Language; >

The following element, Organization, includes basic information about the organization which maintains the information and which has the ability to provide it on request or in response to a purchase order. Basic information is captured using typical names.

<!ELEMENT Organization (Contact*, Name?, Department?, GeographicDomain*, Address?, Telephone*, Fax*, Email*, url*, Availability*, OrderInfo?, CostInfo?) >

GeographicDomain is defined earlier and not redefined here. The GeographicDomain element refers to the jurisdictional responsibility of this organization, e.g., 3rd circuit, not the location of it.

<!ATTLIST Organization ID ID #IMPLIED><!ATTLIST Organization role %OrganizationRoles; >

This creator, implementor – how the organization relates to this doc. The organization’s roles are limited to those defined in the appropriate entity.

<!ATTLIST Organization level %OrganizationLevel; >International, national, state, local, etc., as defined in the entity “OrganizationLevel”.

<!ATTLIST Organization type %OrganizationType; >NGO, etc.

<!ELEMENT Contact (#PCDATA) >

<!ELEMENT Name (#PCDATA) >Name of the organization.

<!ELEMENT Department (#PCDATA) ><!ELEMENT Address (Street?, City?, State?, Zipcode?, Country, Note?) ><!ELEMENT Street (#PCDATA) ><!ELEMENT City (#PCDATA) ><!ELEMENT State (#PCDATA) >

Including provinces<!ELEMENT Zipcode (#PCDATA) >

Including other postal codes.<!ELEMENT Country (#PCDATA) >

<!ATTLIST Country code %CountryCode; ><!ELEMENT Telephone (#PCDATA) ><!ELEMENT Fax (#PCDATA) ><!ELEMENT Email (#PCDATA) ><!ELEMENT url (#PCDATA) >

<!ELEMENT Availability (TimePeriod*, Linkage*) >This element provides information about when the appropriate person to contact is generally available.

<!ELEMENT OrderInfo (#PCDATA) ><!ELEMENT CostInfo (#PCDATA) >

<!ATTLIST CostInfo free (yes|no) "yes" >

<!ELEMENT AccessConstraints (AccessControlInfo?, SecurityInfo?, DisseminationInfo?) >

<!ATTLIST AccessConstraints security NMTOKENS #IMPLIED ><!ATTLIST AccessConstraints allow NMTOKENS #IMPLIED ><!ATTLIST AccessConstraints deny NMTOKENS #IMPLIED ><!ATTLIST AccessConstraints method (positive|negative|other) "positive" >

This is for the user or GLIN or whomever to set up what is to be done. Nmtokens may refer to user names, for example. Under the Dublin Core, this is called “Rights”. I would not use that term her.

<!ELEMENT AccessControlInfo (#PCDATA) ><!ELEMENT SecurityInfo (#PCDATA) ><!ELEMENT DisseminationInfo (#PCDATA) >

<!ELEMENT RecordInfo (Contributor?, LastModification?, NextRecord?, PreviousRecord?, ParentRecord?, ComponentRecords?) >

<!ATTLIST RecordInfo controlID CDATA #IMPLIED ><!ATTLIST RecordInfo originatingControlID CDATA #IMPLIED ><!ATTLIST RecordInfo scheduleNumber CDATA #IMPLIED >

<!ELEMENT Contributor (Name?, Organization?) ><!ATTLIST Contributor OrgID IDREF #IMPLIED >

<!ELEMENT LastModification (#PCDATA) >This is information for management purposes, tracking purposes.

<!ELEMENT NextRecord (DocID) ><!ELEMENT PreviousRecord (DocID) ><!ELEMENT ParentRecord (DocID) ><!ELEMENT ComponentRecords (DocID*) >

If a TOC is not specified, linking can happen through this element. This element refers to “records” rather than “documents” because the conceptual level is computer level, not document level. This element can function as a failsafe for non-TOC docs. In terms of computer overhead, it would be more efficient to use DocID-linkages than TOC linkages, but it also requires more upkeep and programmer input.This element, if used, gives added abilities to move from chapter to chapter or to titles, and up and down.

<!Element DocumentImage (Linkage*)>This element provides a way to link directly to a pdf file or other format of the same law, e.g., MS Word or html.

ELIS/GLIN XML DTD Appendix April 18, 2001 p. A1

<?xml version="1.0" encoding="UTF-8" ?>



<!ENTITY % TitleType "(official|common|citation|other) 'official'" ><!ENTITY % Language "(en|sp|fr|de) 'en'" ><!ENTITY % CountryCode "(us|eu|ca|mx) 'us'" ><!ENTITY % ContentLevel "(1|2|3|4|5|6) '1'" ><!ENTITY % Thesaurus "CDATA #IMPLIED" ><!ENTITY % SubjectTermType "(controlled|uncontrolled) 'uncontrolled'" ><!ENTITY % CitationType "(primary|alternative|universal) 'primary'" ><!ENTITY % PartyActionOnTreaty "(draft|approve|ratify|sign) 'draft'" ><!ENTITY % OrganizationRole "(originator|author|contributor|distributor|

implementor|other) 'other'" ><!ENTITY % OrganizationLevel "(local|state|province|regional|national|

international|other) 'other'" ><!ENTITY % OrganizationType "(government|private|public|ngo|igo|other)

'other'" >

<!ELEMENT LegalDocument (DocID?, Country?, RecordInfo?, LegalDocumentClass?, Title*, Summary*, TableOfContents?, EffectivePeriod?, Subject*, Citation*, CrossReference?, Methodology?, GeographicDomain?, CaseInfo?, TreatyInfo?, PrefatoryStatement?, LegalDocumentText?, Organization*, AccessConstraints?, DocumentImage*) >

<!ELEMENT DocID (#PCDATA ) >



<!ELEMENT LegalDocumentClass (#PCDATA) ><!ATTLIST LegalDocumentClass system CDATA #IMPLIED ><!ATTLIST LegalDocumentClass type CDATA #IMPLIED ><!ATTLIST LegalDocumentClass subtype CDATA #IMPLIED >

<!ELEMENT Title (#PCDATA) ><!ATTLIST Title type %TitleType; > <!ATTLIST Title language %Language; >

<!ELEMENT Summary (#PCDATA) ><!ATTLIST Summary language %Language; >

<!ELEMENT TableOfContents (ContentsEntry*) >

<!ELEMENT ContentsEntry (ContentsEntryText, PageRange?, Linkage?) ><!ATTLIST ContentsEntry level %ContentLevel; ><!ATTLIST ContentsEntry order CDATA #IMPLIED >

<!ELEMENT ContentsEntryText (#PCDATA) ><!ATTLIST ContentsEntryText language %Language; >

<!ELEMENT PrefatoryStatement (#PCDATA) ><!ATTLIST PrefatoryStatement language %Language; >

<!ELEMENT LegalDocumentText (#PCDATA) ><!ATTLIST LegalDocumentText language %Language; >

<!ELEMENT EffectivePeriod (TimePeriod*) ><!ELEMENT TimePeriod (FirstDate?, LastDate?) ><!ELEMENT FirstDate (#PCDATA) ><!ELEMENT LastDate (#PCDATA) >

<!ELEMENT Date EMPTY><!ATTLIST Date Day CDATA #IMPLIED ><!ATTLIST Date Month CDATA #IMPLIED ><!ATTLIST Date Season CDATA #IMPLIED ><!ATTLIST Date Year CDATA #IMPLIED >

<!ELEMENT Subject (Term*) ><!ATTLIST Subject type %SubjectTermType; ><!ATTLIST Subject thesaurus %Thesaurus; ><!ATTLIST Subject language %Language; >

<!ELEMENT Term (#PCDATA) ><!ATTLIST Term id CDATA #IMPLIED >

<!ELEMENT GeographicDomain (place*) ><!ELEMENT Place (GeoName*, (GeoPoint|GeoPolygon)*) >

<!ELEMENT GeoName (#PCDATA) ><!ATTLIST GeoName thesaurus %Thesaurus; ><!ATTLIST GeoName language %Language; >

<!ELEMENT GeoPolygon (GeoPoint*) >

<!ELEMENT GeoPoint EMPTY ><!ATTLIST GeoPoint lat CDATA #IMPLIED ><!ATTLIST GeoPoint long CDATA #IMPLIED ><!ATTLIST GeoPoint alt CDATA #IMPLIED ><!ATTLIST GeoPoint units CDATA #IMPLIED ><!ATTLIST GeoPoint system CDATA #IMPLIED >



<!ELEMENT Citation (Creator*, Title?, Editor?, Publication?, Volume?, Number?, Date?, Publisher*, PageRange?, ParaqraphRange, Note?, Linkage*, URL) >

<!ATTLIST Citation type %CitationType; >

<!ELEMENT Creator (#PCDATA) > <!ELEMENT Editor (#PCDATA) > <!ELEMENT Publication (#PCDATA) >


<!ELEMENT Volume (#PCDATA) ><!ELEMENT Number (#PCDATA) ><!ELEMENT Publisher (#PCDATA) ><!ELEMENT PageRange (first?, last?) ><!ELEMENT ParaqraphRange (first?, last?) ><!ELEMENT Note (#PCDATA) >

<!ELEMENT Methodology (#PCDATA) >

<!ELEMENT CaseInfo (DecisionDate?, Party*, Counsel*, Judge*) ><!ELEMENT DecisionDate (#PCDATA) >

<!ELEMENT Party (Name*, Organization?) ><!ATTLIST Party ID ID #IMPLIED ><!ATTLIST Party role (plaintiff|defendant|other) "other">

<!ELEMENT Counsel (Name*, Organization?, Role?) ><!ATTLIST Counsel Client IDREFS #IMPLIED >

<!ELEMENT CrossReferenceList (CrossReference*) ><!ELEMENT CrossReference (Title?, Citation*, Linkage?)* >

<!ATTLIST CrossReference Relationship CDATA #IMPLIED>

<!ELEMENT Linkage (#PCDATA)><!ATTLIST Linkage Type CDATA #IMPLIED ><!ATTLIST Linkage Code CDATA #IMPLIED >

<!ELEMENT TreatyInfo (DraftedDate?, EffectiveTimePeriod?, TreatyAction*) ><!ELEMENT DraftedDate (#PCDATA) ><!ELEMENT EffectiveTimePeriod (FirstDate?, LastDate?) >

<!ELEMENT TreatyAction (Date?, Party*, ActionDetails?) > <!ATTLIST TreatyAction Type %PartyActionOnTreaty; >

<!ELEMENT ActionDetails (#PCDATA) ><!ATTLIST ActionDetails Language %Language; >

<!ELEMENT Organization (Contact*, Name?, Department?, GeographicDomain*, Address?, Telephone*, Fax*,Email*, url*, Availability*, OrderInfo?, CostInfo?) >

<!ATTLIST Organization ID ID #IMPLIED><!ATTLIST Organization role %OrganizationRole; > <!ATTLIST Organization level %OrganizationLevel; ><!ATTLIST Organization type %OrganizationType; >

<!ELEMENT Contact (#PCDATA) ><!ELEMENT Name (#PCDATA) ><!ELEMENT Department (#PCDATA) ><!ELEMENT Address (Street?, City?, State?, Zipcode?, Country, Note?) ><!ELEMENT Street (#PCDATA) ><!ELEMENT City (#PCDATA) ><!ELEMENT State (#PCDATA) ><!ELEMENT Zipcode (#PCDATA) >

<!ELEMENT Country (#PCDATA) ><!ATTLIST Country code %CountryCode; >

<!ELEMENT Telephone (#PCDATA) ><!ELEMENT Fax (#PCDATA) ><!ELEMENT Email (#PCDATA) ><!ELEMENT url (#PCDATA) >

<!ELEMENT Availability (TimePeriod*, Linkage*) >

<!ELEMENT OrderInfo (#PCDATA) >

<!ELEMENT CostInfo (#PCDATA) >

<!ATTLIST CostInfo free (yes|no) "yes" >

<!ELEMENT AccessConstraints (AccessControlInfo?, SecurityInfo?,

DisseminationInfo?) ><!ATTLIST AccessConstraints security NMTOKENS #IMPLIED ><!ATTLIST AccessConstraints allow NMTOKENS #IMPLIED ><!ATTLIST AccessConstraints deny NMTOKENS #IMPLIED ><!ATTLIST AccessConstraints method (positive|negative|other)

"positive" >

<!ELEMENT AccessControlInfo (#PCDATA) ><!ELEMENT SecurityInfo (#PCDATA) ><!ELEMENT DisseminationInfo (#PCDATA) >

<!ELEMENT RecordInfo (Contributor?, LastModification?, NextRecord?,

PreviousRecord?, ParentRecord?, ComponentRecords?) ><!ATTLIST RecordInfo controlID CDATA #IMPLIED ><!ATTLIST RecordInfo originatingControlID CDATA #IMPLIED ><!ATTLIST RecordInfo scheduleNumber CDATA #IMPLIED >

<!ELEMENT Contributor (Name?, Organization?) ><!ATTLIST Contributor OrgID IDREF #IMPLIED >

<!ELEMENT LastModification (#PCDATA) ><!ELEMENT NextRecord (DocID) ><!ELEMENT PreviousRecord (DocID) ><!ELEMENT ParentRecord (DocID) ><!ELEMENT ComponentRecords (DocID*) >

<!ELEMENT DocumentImage (Linkage*)>

· web viewfor example, the common title “title vii” refers to title vii of the civil rights...

Documents